Stitching images with parallax for naturalness remains a challenging problem. This paper proposes an image stitching method which preserves the flatness of planes in the scene for a natural look. Our method formulates the alignment of images as the camera parameters and the normal vectors of planes. Given a set of feature point matches, a process of grouping points into different layers and rejecting outliers is introduced. According to the epipolar constraint of corresponding points in two images, the focal length and the pose change of the camera are recovered simultaneously. Then, the normal vectors are estimated from the point pairs. To achieve good alignment and guide the warping of images, the model is combined with the mesh deformation as a global similarity constraint. In addition, bundle adjustment is adopted to maintain the consistency for stitching multiple images. Experiment shows that the proposed approach outperforms some state-of-the-art warps on real-world scenes.

This paper is concerned with the template matching problem and a weighted smallest deformation similarity measure based on the nearest neighbor (NN) matching of points is proposed. The smallest location distance between each point in the template and points matching it in the target image is employed to penalize the deformation explicitly. To reduce the negative effect of points that belong to the background in the template, the weights are added to these points by counting the number of their NN matches of the points around the target window. The proposed similarity measure is robust to occlusions, background outliers and complex deformations. Experiment shows that the proposed method improves the state-of-art performance on real-world scenario benchmarks and can be applied in rough positioning of surface mount technology (SMT) components.

Homography estimation is a fundamental task in computer vision that involves obtaining the transformation between multi-view images for image alignment. Although convolutional neural networks (CNNs) have shown state-of-the-art performance in this task, few works have explored the use of transformer-based models, which have demonstrated superiority in high-level vision tasks. In this paper, we propose a strong baseline model for homography estimation that combines a swin transformer feature representation for global features and a CNN feature representation for local features. Additionally, we introduce a cross non-local layer to coarsely search for matched features within the feature maps. In the homography regression stage, we adopt an attention layer to drop out weak correlation feature points from the channels of the correlation volume. Our experiments show that our method outperforms the state-of-the-art methods in 8 Degree-of-Freedoms (DOFs) homography estimation. Code is available at https://github.com/mingxiaohuo/ABHE.

Template matching in unconstrained environment with complex deformation, occlusion, and background clutter is a challenging task. Recently, some measures which are robust to outliers were presented, however, they fix the window size and thus cannot handle large-scale change. In this article, a multiscale template matching method based on nearest neighbor (NN) search is proposed. To discover the effect of scale to the measure, the expectation of the diversity similarity (DIS) is derived by probabilistic analysis. Then, a scale-adaptive measure is provided by extending DIS and penalizing the deformation explicitly. Moreover, for rectangular template, weights are appended to points to suppress the negative effect of background pixels, and for masked template, foreground pixels are separated from the candidate window based on NN field. In addition, a scheme for preselecting the candidate positions of object detection is given. Experiment on the real-world scenario benchmarks and surface-mount component (SMC) positioning shows that the proposed method is robust to scale changes along with other challenging aspects and outperforms the state-of-the-arts using both color and deep features.

Alignment for natural images in unconstrained environment is a challenging task. Despite the success for complex deformation, existing feature-based methods may be confused in low-textured regions where features are insufficient, while pixel-based approaches may fail when color changes. In this paper, a parametric chamfer alignment method based on mesh warping model is proposed. Warped positions of mesh vertices are considered as parameters and estimated by optimizing an object function, which measures the chamfer distance of edges and the smoothness of warping. To distinguish the sharpness of pixels, edges are detected through K-means cluster and weights are attached to different levels of edge points. In addition, after the warping model is initialized by feature-based alignment, a growing technique for registering the vertices is presented. Experiment shows that the proposed method outperforms some state-of-the-arts on real data and can be applied in stitching ceramic sanitary ware images.